Abortion policy implementation in Ireland: successes and challenges in the establishment of hospital-based services

Abstract

Objective

To describe successes and highlight remaining challenges in the establishment of hospital-based abortion services after legal change in the Republic of Ireland.

Methods

We conducted a mixed-methods study on the implementation of abortion policy in Ireland. In this manuscript, we present the results from a qualitative analysis of in-depth interviews conducted with hospital-based providers, service users, and key informants. We used Dedoose software to conduct a thematic analysis of the data.

Results

We report findings from interviews with 28 obstetrician gynecologists, midwives, psychiatrists, anesthesiologists, and nurses; a subset of 7 service users who sought care in hospitals; and 27 key informants. In this analysis, we describe how key themes that pertain to information, capacity and power, facilitated and hindered the implementation of hospital-based abortion services. We found that individual champions are key to establishing the service, but their motivation is not always sufficient to integrate abortion into existing clinical services, and conscientious objection is a persistent barrier to expanding abortion services. The main challenges highlighted here are lack of abortion provision at some hospitals and limited access to surgical abortion at most hospitals due to provider-level, logistical, and infrastructure barriers.

Conclusions

This study presents new information on how abortion policy is implemented on the ground in hospital settings. Its findings can inform public health officials and providers in Ireland and other countries wishing to establish abortion services.

1. Introduction

In 2018, following a popular referendum, the Republic of Ireland significantly expanded the grounds for legal abortion (Bardon, 2018). Termination of pregnancy (TOP) was previously banned under all but the most extreme circumstances. It is now legal without restriction as to reason under 12 weeks, and beyond that if the woman's life or health is at risk, or if there is a fatal fetal abnormality (FFA) likely to lead to the death of the fetus within 28 days of birth (Oireachtas, 2018). This legal change followed years of advocacy by women's, civil society, and physician advocacy organizations (Enright et al., 2015; Bergen, 2022). Provision of services started in January of 2019. Abortion is provided free of charge through the public health system, and the care model emphasizes community provision of early medical abortion (EMA) by general practitioners (GPs). Hospital maternity units are responsible for providing abortions beyond nine completed weeks, including those performed in the setting of FFA (Mullally et al., 2020). A Department of Health report published in June of 2021 shows that 6577 abortions were reported, 98% of which were performed under Section 12 (early pregnancy) (An Roinn Sláinte, 2020).

This change represents significant progress, but new laws do not seamlessly lead to accessible abortion services. In many countries, access to abortion remains limited years after legal change (Trueman & Magwentshu, 2013; Chavkin et al., 2018). Financial barriers are common, as are other obstacles created by abortion opponents to restrict access to legal abortion (Cohen & Joffe, 2020). Many forces are at play as abortion is inherently a health and socio-political issue. Early reports of the implementation of abortion services in Ireland suggest that community-based care provided by GPs is meeting the needs of many service users, but that referral pathways into secondary care are not always smooth (Mishtal et al., forthcoming). Further, only 10 out of the 19 existing maternity hospitals are currently providing abortion services (Mullally et al., 2020). At one tertiary maternity hospital, the service relies on a small number of consultants, which raises concerns about its sustainability (O'Schaughnessy et al., 2021). A survey of medical and nursing staff at the same hospital revealed that most respondents (88%) had not received clinical training prior to the implementation of TOP services, and that most (94%) wanted to receive more training (O’Shaughnessy, Leitao, et al., 2021). A qualitative study of ten fetal medicine specialists from five of the six Irish fetal medicine units described challenges in providing care for patients with fetal anomalies that do not fit the strict criteria set forth by the legislation (Power et al., 2021).

These reports highlight some of the barriers individual hospitals and clinicians encountered in establishing TOP services in Ireland. Our team undertook a mixed-methods study to examine the process of implementation of the new abortion policy across Ireland and through the viewpoints of multiple stakeholders. We use in-depth interviews with key informants, service users and hospital providers across Ireland to identify barriers and facilitators in this process. Here, we focus on the establishment of hospital-based TOP services.

2. Methods

In this mixed-methods study we relied on three types of data: (1) semi-structured interviews with key informants and target groups, (2) quantitative records collection from key actors, and (3) a desk review of public policy documents. For this paper, we analyzed semi-structured interviews of key informants and hospital-based providers throughout Ireland, as well as interviews from service users who sought care in hospitals. Further details on the entire study design are published elsewhere (Mishtal et al., forthcoming). Here, we report our findings in adherence with the Consolidated criteria for the reporting qualitative research (COREQ) checklist (Tong et al., 2007).

2.1. Research team and reflexivity

The research team included 11 researchers in the fields of social science, public policy, medicine and public health. The team was led by an anthropologist (JM) and included other anthropologists, as well as physicians and public health experts across four countries (Ireland, United Kingdom, United States, Switzerland). BMS (MD, MPH, obstetrician-gynecologist) and JM (MA, PhD, anthropologist) conducted the interviews with hospital-based providers; DC (PhD Candidate, sociology) and LG (PhD, historian) conducted those with service users; and JM conducted those with key informants. All interviewers have extensive formal training and several years of experience with qualitative data collection methods. None had any established relationships with study participants prior to the interviews. Although interviewers did not discuss their own views about abortion explicitly, it is possible that participants could have perceived an explicit bias in support of abortion provision based on the interviewers’ research interests and questions.

2.2. Study design

2.2.1. Theoretical framework

We used the Contextual Interaction Theory (CIT) to guide this work. This is a policy implementation theory that has as its basic assumption that “the course and outcomes of the policy process depend not only on inputs, but more crucially on the characteristics of the actors involved, particularly their motivation, information, and power” (Bressers, 2004). The theory also focuses on the interactions of actors within the implementation process and highlights that “policy instruments cannot be seen in isolation from the circumstances in which they are applied” (Bressers, 2004; Bressers, 2011). CIT aims to predict different types of process interactions based on varying combinations of the three core variables of motivation, information and power (Bressers, 2004). We chose CIT because the abortion policy implementation process involves activities and interactions between several actors (individuals and/or organizations) who work together in a variety of contexts, including healthcare providers, government, and non-governmental organizations. A predictive policy implementation model can identify factors that affect successful implementation, which could be corrected early in the implementation process.

2.2.2. Participant selection, recruitment & setting

We used purposive sampling to select participants for the study (Bernard, 2006). Hospital-based providers were obstetrics & gynecology trainees and consultants, midwives, nurses, and other healthcare providers working in maternity hospitals throughout Ireland. We recruited providers from all Irish maternity hospitals, including those that do not provide abortion services. There are 19 maternity hospitals in Ireland, all of which are public. However, private care is also available in the public hospitals. Midwives in Irish hospitals work alongside consultants and trainee doctors and provide obstetric and gynecologic care. They also provide perinatal bereavement services, ultrasound scans, and manage patients admitted for medication abortion. Nurses and midwives are not permitted to perform uterine aspiration procedures, but nurses are typically involved in scheduling and supporting surgical cases.

Key informants for this study were public health and medical professionals who contributed to the implementation process in leadership roles, and representatives of reproductive health advocacy organizations. We recruited respondents via e-mail utilizing established networks of provider contacts as well as snowball sampling. In our interview invitation email, we invited providers to participate in the study and offered the option to schedule in-person or virtual interviews.

We recruited service users, women 18 years or older who sought abortion services in Ireland in 2020, through flyers distributed in GP offices and advertisements posted on social media platforms (Facebook, Instagram, Twitter). All participants provided informed consent prior to beginning the interview. We de-identified all interviewees except for those who explicitly chose to be named: the former Minister for Health Simon Harris and the former national clinical adviser to the Health Services Executive (HSE) for abortion service implementation Peter Boylan.

2.2.3. Data collection

CIT guided our development of in-depth interview guides centered around the theory's three core variables: motivation, information, and power (Bressers, 2004). We also used previous work on the implementation of abortion services and our clinical experience to inform specific questions within each variable domain (Chavkin et al., 2018).

We conducted all interviews, virtual or in-person, in private locations, with no one else present but the interviewer and respondent. Interviews took between 60 and 90 ​min to complete. A professional transcription service transcribed all the interviews, which we recorded using a digital tape recorder. We also took notes during and after each interview and shared them with the research team. We did not return interview transcripts to participants for comment and/or correction but reviewed the data at three virtual dissemination meetings with Irish public health officials, nonprofit representatives, healthcare providers, and other stakeholders. These meetings offered participants an opportunity for discussion and a venue to provide feedback on the research process and findings.

We stopped recruiting participants once the research team agreed we had reached thematic saturation with our samples (Saunders et al., 2018).

2.3. Data analysis, reliability, and validity

Five researchers (BS, JM, KR, DC, LG) coded and analyzed interview transcripts using Dedoose software (Dedoose, 2018). Two coders independently coded 20% of the transcripts and met to discuss codes, ensure coding consistency, and define a codebook. We discussed and resolved any discrepancies with a third researcher, with no significant disagreements in the interpretation of the emerging data. We further defined the codebook and discussed analysis during periodic team meetings. The coding process used both open and axial coding, in line with grounded theory's “dynamic and fluid process” which includes both predetermined codes and those that emerge from the data (Coffee & Atkinson, 1996; Strauss & Corbin, 1998).

Upon completion of coding, we organized findings in a data display structured according to the CIT core variables of motivation, information, and power. We then articulated themes for each CIT variable, and considered whether each theme acted as a facilitator or barrier in the context of the implementation of hospital-based abortion services. Here, we present findings in narrative format, describing the initial implementation steps, the successes and remaining challenges.

We took several steps to ensure the reliability and validity of our findings (Morse, 2015). These included double coding, sustained memo writing throughout the data collection and analysis phases, and discussions of subjectivity among members of the research team. We also engaged participants and other stakeholders in research dissemination meetings and offered an opportunity to discuss the findings prior to proceeding with manuscript writing and publication.

2.4. Ethics & funding

The UNDP-UNFPA-UNICEF-WHO-World Bank Special Programme of Research, Development and Research Training in Human Reproduction (HRP), a co-sponsored program executed by the World Health Organization (WHO), funded this work. The WHO’s Ethics Review Committee approved the study, as did the University of Central Florida’s Institutional Review Board.

3. Results

We included interviews with 28 hospital-based providers, 27 key informants, and 7 service users in this analysis. Table 1a, Table 1b, Table 1ca, 1b and 1c show the characteristics of the interviewees (see Table 1a, Table 1b, Table 1c). Table 2 shows a summary of key facilitators and barriers encountered in the establishment of hospital-based abortion services in Ireland, with their relevant contextual interaction theory variables.

Box 1. Abortion Policy Implementation in Ireland: Summary of Key Lessons for Other Countries.

• •Implementers should consider access to surgical abortion early in the planning of abortion services, as offering a choice of abortion method is an important aspect of providing quality care. Provider-level and logistical/facilities-level barriers may emerge. Specific training efforts for different provider types and expanding capacity for outpatient aspiration procedures are possible solutions.
• •Political will is key to ensuring successful implementation of abortion services. However, it must be accompanied by swift action and ministerial leadership to avoid delays in local guideline/protocol development and on-the-ground implementation. Planning for service implementation after legal change should ideally take place before the law comes into effect.
• •Implementers should devise strategies to address conscientious objection from providers and ancillary staff. These may include staff recruitment protocols and concerted training efforts. Increasing exposure to abortion during medical and nursing training is one way to create a pipeline for future service providers. Clinical leaders should clarify with all staff the roles and ethical obligations of conscientious objectors in accordance with local laws and regulations.
• •Values clarification workshops are helpful for willing providers and conscientious objectors alike and should be considered an important component of abortion policy implementation.

Alt-text: Box 1

Table 1a.

Descriptive characteristics of 28 healthcare providers who participated in in-depth interviews on the implementation of hospital-based abortion services in Ireland.

	n (N ​= ​28)
Provider Type
Obstetrics & gynecology consultant	10
Obstetrics & gynecology trainee	2
Midwife (manager/director)	7
Midwife staff	4
Anesthesiologist	2
Nurse	1
Psychiatrist	2
Hospital type
Provides abortion services	24
Provides only for FFA	2
Does not provide abortion services	2
Countya
Leinster	16
Connacht	8
Ulster	1
Munster	6
Involved in policy
Yes	10
No	18

^aSome providers work in more than one province.

Table 1b.

Descriptive characteristics of 7 service users who sought abortion services in hospitals and participated in in-depth interviews about their experiences.

	n (N ​= ​7)
Age (mean, years)	37.7
County
Dublin	3
Waterford	1
Donegal	1
Cork	1
Meath	1
Grounds for TOP
Under 12 weeks	1
Fatal fetal abnormality	3
N/A – did not have TOP in Ireland	3
Ethnicity
White Irish	7

Table 1c.

Key informants by category.

	n (N ​= ​27)
Health Services Executive or government representatives	13
Representatives from reproductive health and justice organizations	10
Representatives from organizations involved in care provision and advocacy	4

Table 2.

Summary of key facilitators and barriers encountered in the establishment of hospital-based abortion services in Ireland, with their relevant contextual interaction theory variables.

Facilitators	Barriers
Development of guidelines using existing guidelines from other countries •Information	Delay in developing clinical guidelines •Capacity/power – limited top-down leadership
Establishment of collaborative, multidisciplinary group of providers •Information – sharing knowledge among providers •Capacity/power – building capacity to provide care	Limited/delayed preparation on the ground in hospitals •Capacity/power – limited top-down leadership
Motivated champions step up to provide service •Motivation – individual provider motivation as a catalyst to initiate services	Less emphasis on secondary care compared to community-based care •Capacity/power – limited top-down leadership
Many providers have existing clinical skills (from working abroad or from miscarriage care) that are useful in abortion care •Information •Capacity/Power	Minimal specific clinical and logistical training •Capacity/Power
Values clarification exercises help clarify roles •Information •Motivation	No formal recruitment of willing providers •Information – lack of open discussion •Capacity/power – operating in context of stigma
Providers find TOP work meaningful •Motivation	Conscientious objection among providers and ancillary staff •Motivation
Supportive environment for providers •Capacity/power – providers have peers who provide source of support •Motivation – providers feel supported in their decision to provide TOP •Information – providers can share clinical information & expertise	Workload/convenient objections among providers and other staff •Motivation •Capacity/power – limited staffing, workload burden
MyOptions helpline effectively informs and directs users towards TOP services •Information	Referrals to secondary care dependent on individual coordinators •Capacity/power
Timely referrals from primary to secondary care •Information •Capacity/power	Inadequate facilities/insufficient space as a barrier to establishing TOP care •Capacity/power
Exposure to abortion care normalizes it and decreases objections •Information	Services marginalized to evenings and weekends •Capacity/power
Use of ambulatory procedural space to provide manual vacuum aspiration (MVA) •Capacity/Power	Inadequate facilities/insufficient space as a barrier to providing surgical TOP •Capacity/power
	Provider hesitations as a barrier to providing surgical TOP •Motivation

3.1. Beginnings of a secondary care service

3.1.1. Physicians take the lead in guideline development

After the referendum to Repeal the Eighth Amendment took place, the then Minister of Health Simon Harris felt strongly that “Irish women had waited long enough,” and that the service should be implemented by January 1st of the following year. In this six-month period, physician respondents described scrambling to develop clinical guidelines in preparation for service implementation, with little guidance from the HSE. One provider who was involved in this process felt that most implementers were focused on community-based provision, while “nobody really gave any thought to [secondary care] and [nobody bothered] to do anything about it and still haven't, to a large extent actually.” [Provider 48, OB-GYN]

Another provider explained that several healthcare providers founded a multidisciplinary group called the Southern Task Group for Abortion and Reproductive Topics (START) to share information and collaborate in establishing abortion services. This group included at that time OB-GYNs, GPs, and a psychiatrist, and its members drafted a position paper soon after the referendum. A few OB-GYNs who were members of the START group later wrote official clinical guidelines for the Institute of Obstetricians & Gynecologists. Several respondents, however, felt that this impetus came late, and that as a result, the process was rushed. On January 1st, when the services were scheduled to start, the guidelines were still being formalized. As one respondent explained:

“As far as we could see, there was no one else really at the HSE or Department of Health level working on this and that was kind of fascinating to us. It was like they were all, ‘Yay, yay, it's all been repealed’. It's like, ‘Ok, guys, but what is the service going to look like?’ […] I think was 3rd November […]. No guideline, no model of care, no heads of bill, no further along’ […] every single night from that day, so I worked as normal and then I would come home every evening and you know, read someone's national guideline, work on the text […].” [Provider 21, OB-GYN]

Physicians who were involved in drafting clinical guidelines explained that some points were initially controversial, such as whether all women would be required to undergo ultrasound scanning or Rhesus (Rh) typing (neither were ultimately required). In this process the guideline writers found that reviewing existing guidelines from other countries was helpful. As Provider 21 put it: “we took American guidelines, Canadian guidelines, Dutch, French, English, Norwegian, anything we knew we could get online and just started working out, well, what were the doses? […]”. Several respondents also mentioned a beneficial experience, whereby abortion service implementers from other countries and representatives from the World Health Organization came together in Ireland to share experiences.

3.1.2. Motivated champions step up to provide abortion care

While preparations were underway in terms of clinical guideline writing, respondents described few efforts to prepare for service provision on the ground in individual hospitals. Several providers from different hospitals felt that leaders were in denial about the upcoming abortion service requirement and postponed planning for it, as one OB-GYN explained:

“I was writing, writing, writing [guidelines] and yet I could see that there was no work happening on the ground to prepare the maternity units, the hospitals for preparing the service. There was this amazing kind of feeling of oh yes, that's going to happen somewhere else. Somewhere else, someone else, some other doctor, some other where, some place somewhere that I don't have to think about that. It's going to happen.” [Provider 21, OB-GYN]

A midwife manager in a different hospital was one of several respondents who also described a lack of open discussion among leadership and providers about who would be willing or unwilling to provide the service:

“Nobody seemed to want to engage in conversation [to say] yes I will, no I won't, you know, there was no clear kind of opinion on it, or position on it. So, it actually took quite a while to get clarity as to who might potentially offer their service, and I think discretely people came and said, well actually yes I would have had experience of this. So there were two consultants eventually that said, well potentially yes this would be something that I would consider offering and assisting with.” [Provider 47, midwife manager]

Several interviewees said there were no open conversations regarding who would provide TOP because abortion remains highly stigmatized. In this context, individuals who were motivated to provide abortion were recruited informally or took the initiative to become involved. Midwives' and OB-GYN consultants’ willingness to step up emerged as a key determinant of whether services were established at individual hospitals. One midwifery manager explained that a successful strategy for recruiting providers was to have “targeted discussions with a handful” of midwives whom she had identified as “champions” that would “potentially be prepared to lead the service” [Provider 41, midwifery director]. Another midwife explained how she volunteered for a midwife coordinator role just before the service provision was scheduled to begin:

“[I came on board] just before […] the Christmas holidays. We knew it was happening. We knew repeal the 8th had happened and it was going to be provided in hospitals from January 2019 and I was interested in helping with that and it was just, you know, an interest of mine. So I went to my director of midwifery and said, ‘Look, you know, you haven't advertised anything, but if you were looking for someone to help coordinate from a midwife's perspective, I would be interested’. So, she said yes, fine, that's grand.” [Provider 13, midwife]

Several midwives and physicians who volunteered were motivated by their previous experiences with women unable to access this care. One obstetrician-gynecologist explained:

“I just found the courage to legitimize abortion care from all of those years of hearing those stories and thinking, ‘That is so, so wrong’. You know, I remember sitting with women and saying, ‘Well, look, what are you going to say?’ So, they'd ask me like, ‘How can I make this look like it was a miscarriage?’ […] Like it was just horrendous, absolutely horrendous for her, you know […] Yes, it was mental. It was absolutely mental.” [Provider 31, OB-GYN]

Some providers also described long-standing “pro-choice” sentiments as additional motivators to become involved in the service, as well as an interest in new opportunities. One midwife explained:

“I would have always been pro-choice and you know, I've been a nurse, I've been a midwife, I'd worked a long time in the emergency room here in the maternity hospital […] And then when I heard about all this, it was a new service, a new opportunity […] which is why I put myself forward and oddly, I was the only person who did, you know.” [Provider 13, midwife]

Finally, a few providers were motivated by a sense of professional obligation or from experiences working in countries with more expansive access to abortion.

3.1.3. Conscientious and convenient objections from providers as an early barrier to establishing services

While motivation from individual providers was key to establishing abortion services, lack of willingness to participate took many forms. The most cited reason for reluctance to participate was conscientious objection (CO), which emerged early in the implementation process and was present to varying degrees in all hospitals. CO was an issue among consultants and midwives but also among other staff such as operating theatre nurses, unit nurses, and anesthesiologists. One midwife explained that objectors sometimes caused burdens to willing staff:

“We're able to run the service, but [sometimes there is] a conscientious objector on shift who basically just blocks everything completely. So there was a night that I was on day shift, which is usually half seven till half eight and I had to stay in the hospital till 11 o'clock because the midwife that was coming on for night shift was a conscientious objector and refused to even go into the room. So I had to stay […] there were a few issues like that the conscientious objector wouldn't answer an emergency bell and stuff, so those were kind of barriers to it being well-staffed […]” [Provider 37, midwife]

While this midwife felt that this kind of obstruction was rare, and that it did not impede service provision in her hospital overall, in other hospitals where a majority of providers claimed conscientious objection, services could not be established:

“Most hospitals are providing. We still have a few because we have 19 maternity hospitals and legally, they're all supposed to be providing, but some of the smaller ones are not because the consultants and management are conscientious objectors.” [Provider 13, midwife]

Several providers explained that, because abortion was not openly discussed, identifying and categorizing objectors was difficult. Further, there were no centralized or systematic ways to regulate CO. The Health (Regulation of Termination of Pregnancy) Act 2018, the Institute of Obstetrics & Gynaecology's Interim Clinical Guidelines for termination of pregnancy and the Irish Medical Council's Guide to professional conduct and ethics all allow for CO, but state that objectors must refer patients to other providers and must provide care in an emergency. Neither the legal nor the guideline documents require any notification or registration of conscientious objectors (Oireachtas, 2018; IOG, 2018; Irish Medical Council, 2019). Former Minister of Health Simon Harris explained that CO is handled at the level of each hospital, where managers must ensure they have “enough clinicians on [their] staff who will provide the service” [KI27]. He emphasized that institutions cannot be conscientious objectors and that a solution to CO would be for hospitals to recruit consultants specifically to perform TOPs, an approach which was helpful in some hospitals.

Many respondents also explained that other types of objections had also emerged as barriers to service provision - these were a matter of convenience or workload but were sometimes disguised as ethical objections. One anesthesiologist explained:

“They were signing the conscientious objection forms because so many of their colleagues were refusing to perform procedures on moral grounds that a lot of extra responsibility was being placed on them to come in earlier and to carry all that workload and they didn't feel that it was fair.” [Provider 30]

3.1.4. Initial fears prove unfounded

Another initial barrier that emerged in some interviews around the concept of motivation or willingness to provide TOP was fear. Several respondents explained that some healthcare providers were influenced by fears of the unknown; fears of being overwhelmed by the number of cases; fears of complications; fears of malpractice suits; fears of having protests outside hospitals that provide abortions; and fears of being stigmatized or of damaging their careers through an association with abortion. Several providers mentioned reluctance to be seen as “full-time terminators” [Provider 26, OB-GYN] or “TOP technicians” [Provider 41, midwife director]. A midwifery manager, for example, described the case of a midwife who “didn't want to be kind of perceived professionally as […] the TOP midwife […], but wanted to be a midwife who works in a service where TOP is available” [Provider 41, midwife director]. In most cases, however, respondents described that providers' initial fears proved unfounded. For example, complications were rare, and none of the providers mentioned actually feeling personally stigmatized. Most providers said protests were minimal where they did occur, and quickly dissipated after the initial implementation period. Finally, the demand for TOP services was not as high as some feared, as one midwife explained:

“I suppose there was also this worry at the beginning of the service that we were going to be doing so many abortions, that we'd have to curtail other services, and none of that actually ever materialized and we're well capable of managing the caseload that comes through.” [Provider 23, midwife]

3.1.5. Providers learn on the ground and use existing skills

While most respondents said they did not receive any formal clinical trainings on TOP, midwives and obstetricians generally felt that they already possessed many of the technical skills required to provide medical and surgical abortions. Even providers who had not previously performed abortions felt that the skills acquired in managing miscarriages were helpful for TOP. However, even some of the most technically skilled providers said that they would have preferred to receive TOP-specific training, as they were unfamiliar with the psychosocial, legal and logistical aspects of TOP care, as one midwife explained:

“Literally, when I started in January, I got handed a folder with a bunch of information on all the legislation, on the consent forms and everything. So I literally had to do a lot of reading on all the documents I had to see what the guidelines were. I kind of had to educate myself on that. And then everything else I suppose because of my age and my experience, you know, I'm able to communicate with the ladies […] But it was a different way of speaking. Like if you go in to a lady with a miscarriage, it's her baby. You go in to a lady with a termination, it's her fetus. So, it's worked well, but like all the training really has been informal. [Provider 13, MW]

An OB-GYN said that while some staff had worked abroad and were skilled in TOP, others who were younger “had no idea” [Provider 48]. An anesthesiologist added:

“There was no training at all, and I suppose whilst the procedures themselves, like certainly from an anesthesia management point of view, wouldn't be particularly challenging and wouldn't be hugely different from many other procedures we would be doing, I think that people definitely need training around how to communicate with patients and maybe to understand the social circumstances and medical circumstances in which people present for surgical terminations.” [Provider 30]

A handful of providers said they had chosen to attend formal training (such as manual vacuum aspiration trainings) organized by outside organizations and found it helpful, but this was rare.

3.1.6. Values clarification exercises help define provider roles

While formal trainings in clinical and technical skills were rare, many respondents did participate in a series of values clarification workshops organized prior to the rollout of the service, which they almost universally described as highly valuable. Some respondents also participated in other workshops and a handful trained as facilitators led similar workshops through the Institute of Obstetrics & Gynaecology. One key informant explained that almost 200 people from different sites attended a workshop set up in a train-the-trainer model [KI25]. The workshops allowed staff to reflect upon their beliefs, and potential objectors articulated specific elements of abortion care that they would or would not be willing to participate in, as one provider explained:

“Doing the values clarification workshops took a lot of the pressure off because staff realized they could object, that they didn't have to do it, they didn't have to be in theatre, they didn't have to do anything, but then that it was ok for them to recover a patient or it was ok for them to check them in. So suddenly, the pressure was off.” [Provider 24, OB-GYN]

Some providers added that the workshops are helpful even for those who are already willing to provide the service; one OB-GYN felt that the workshops helped her process emotions that arose, for example, while performing surgical TOP cases [Provider 24]. Another OB-GYN was also surprised to find the workshops helpful:

“Values clarification workshops are excellent and I participated in one, even though I thought my values were pretty clarified. But I participated in one […] and it was excellent and I almost wasn't going to go because I thought it wasn't for me, but it was a really, really way of positioning people. So if you're a theatre nurse and, you know, you don't want to be involved, so you know, when you say you don't want to be involved, does that mean you don't want to be present for a surgical termination? Are you happy to set out a theatre trolley? Are you happy to bring in an ultrasound machine? Are you happy to be involved in her anaesthetic? And often when you tease out these small things, they're actually ok with all of that.” [Provider 22, OB-GYN]

Dr. Peter Boylan, the national clinical adviser to the HSE for abortion service implementation felt that a positive aspect of the workshops was that “there were some people at the courses who actually changed their mind afterwards and said yes [to providing abortions].” [KI 20]. A few other respondents, on the other hand, felt that the workshops cannot “influence value sets that are already there” [Provider 49, MW director]. One provider felt that the workshops softened the stance of some objectors and facilitated the introduction of the service:

“I think it's a positive effect on people and we had the managers, for example, in the maternity ward, and at the beginning, they were not open and they didn't want the service to be brought to the maternity at all. And after they participated in the workshop, even if they said that maybe they're not 100% towards, they are still open for the discussion and you know, trying to facilitate with finding staff that is willing to look after these women. So we were able to at least facilitate the service in the ward and they were willing to find the staff and you know, open to the discussion.” [Provider 42, OB-GYN]

Respondents also mentioned that multiple workshops need to be organized and at different times to reach a wide range of staff, and that this should be done during staff working hours. Some felt that workshop organizers had targeted obstetricians, but that ancillary staff such as nurses and theatre managers were equally important. Some added that one challenge in reaching staff can be that those who are “struggling the most with [the issue]” are least likely to attend the workshops [Provider 40, MW director]. Repeating workshops after a year or two was cited as helpful to address new issues and reach new staff.

3.2. A working service for most users, slowly integrated into hospital routines

3.2.1. Meaningful work and supportive environment as abortion becomes normalized

Respondents who are abortion care providers explained that they feel good about their work and are proud to be providing an important service for women. A nurse who is a theater manager also said: “I really feel that we're supporting the women and I have really felt a great sense of pride that I am involved in the service” [Provider 25]. A midwife added:

“The women are great and it's great to be able to provide, like when you can see the tension, stress, the upset, you know, how much it depends on this, like you know. And then the fact that you can help them and make such a difference is wonderful really.” [Provider 13, MW]

An OB-GYN explained that abortion care is a “human right” and that she is “ethically, morally and socially content with providing” it [Provider 21].

Despite initial fears, no interviewees mentioned personal experiences with being stigmatized for providing abortion. Several providers reported that they felt supported by hospital management and by other providers - both within the same hospital and across the country. Some OB-GYNs and midwives joined the START group, an interdisciplinary group of abortion providers which comprises mostly general practitioners. The group was instrumental in developing clinical guidelines but also provides a source of advice and “collegial support” [Provider 36], as one midwife explained:

“We all work really well together and we also have a WhatsApp group […] the advice everyone gets from each other, from the support from other GPs. You'd often have one of the consultants who would jump in and would advise. Like in the medical community, [abortion care] is very well supported. Like all our peers are supporting us, as in anyone who's involved in the service offers great support to everybody else involved in the service.” [Provider 13, MW]

Several providers also reported that with time, abortion services started to be considered as a routine service provided in the hospital. One consultant OB-GYN from a providing hospital explained that “within a few months of being introduced [TOP became] a reasonably normal or accepted part of the service” [Provider 43]. Another OB-GYN added that once the novelty of the service diminished, it “became just one of those things” that people “don't even talk about any more” [Provider 21]. Several felt that staff attitudes had also improved over time, even among those who initially objected to the service. One midwife explained the case of a Catholic nurse who had initially been very critical of TOP patients and whose outlook changed as she became exposed to the patients' stories and personal situations:

“I've seen it myself with my own eyes, when people are around TOPs more, they get more experience in it. They just become so much more adjusted to it and as I said, they realize it's not a big procedure. It's safe and you know, when you speak to a woman one on one, you understand a lot more her whole situation.” [Provider 37 MW]

One OB-GYN even felt that the number of conscientious objectors decreased over time as people are “more accepting of the normality and the routine nature of termination of pregnancy” [Provider 26]. Dr. Peter Boylan [KI21] also added that objections to the introduction of surgical services such as manual vacuum aspiration (MVA) also decreased over time as people saw that the service was “not as bad as they thought it was going to be […] it's another procedure”.

3.2.2. Perception of quality service for users who access it

Our data show mixed reviews on the accessibility of hospital-based TOP services, but overall satisfaction with the services once accessed. Service users who were able to obtain abortions in hospitals were generally pleased with the care they received from individual providers. Several of them described “lovely midwives” and said they received “fantastic care” from consultants, though two highlighted that they felt particularly lucky because they were private patients. All reported timely and appropriate referrals from primary to secondary care. However, at least one patient expressed frustration from having to travel to Dublin several times during the work-up of a fetal anomaly, and to another hospital for the TOP as her local hospital did not provide. Although the travel “took a toll on [her] emotionally,” she received “fantastic care” as the hospital team “went above and beyond to care for [her]” [Service User 28].

Several service users in our sample were unable to access abortion in hospitals because they had fetal anomalies that did not meet the strict prognostic criteria for legal TOP in Ireland, and fetal medicine specialists highlighted this as an important barrier. Data from both service users and providers showed that the 12-week gestational age limit was a barrier to accessing hospital services, even for those who were close to but not beyond it. We acknowledge that these barriers are important but do not discuss them here as they will be discussed in detail in a separate manuscript [23].

Most respondents from hospitals that provide TOP felt that, despite some initial challenges in establishing it, the service is working well and meeting the needs of most users. Most also added that they were able to maintain service provision with minimal impact from the COVID-19 pandemic. One midwife was among several providers who said patients are giving “positive feedback” about the service, while “there haven't been any complaints” about it [Provider 33, midwife]. Several providers felt that this is in part because the number of abortion patients has been lower than expected, thus allowing hospitals to meet the demand despite some of the remaining challenges. Some also explained that remarkable progress has been made in comparison to the pre-referendum absence of services:

“Well I think from where we were two years ago even, we've come a huge way […] women are getting better care. Women are not travelling across to the UK for terminations as much. It still happens, but it's definitely not as much. So therefore, they are getting better pre and post care by the person who's actually carrying out the procedure. But I think we've come a long way and of course everything could be improved, and it probably is being improving as we develop the service, so I think we've done a good job overall. Yes.” [Provider 50, MW]

At the HSE level the overall perspective on the hospital service was similarly positive, as former Minister of Health Simon Harris explained:

“Even though we don't have the service provided in every hospital today in Ireland, we have largely ensured that it is provided in enough places in Ireland. Now, that's not to say I wouldn't like to see it provided in more hospitals, I personally would, but I think the effort at erecting that barrier largely failed, because there were enough Doctors willing to stand by women and respect the law and the democratic will of the people. So I think that was one very large potential barrier that has been overcome.” [KI 27]

In hospitals that do not provide or within hospital networks where some hospitals provide while others do not the prevailing opinion was also that there are enough providing hospitals that those who need TOP can access it, albeit not always at their local hospital.

Key informants and hospital providers alike explained that they believed that Irish women are well informed about the services available. Although hospitals that provide TOP services do not advertise them, women who are seeking TOP can contact a government-run phone and Internet helpline (MyOptions) that connects them with GP offices and family planning clinics that provide abortion. Most of the service user respondents did mention ringing MyOptions at some point in their journey. Women who need secondary care are typically referred to providing hospitals by the GPs who first see them. One midwife explained:

“All the women seem to have the information and how to access it […] My Options gives them a lot of information and even though we have the information here, when the women come in to us, they all meet with a GP first and they have been counselled and they've received all the information, so they're very well informed when they actually reach me, you know, which is fantastic really. [Provider 13, midwife]

Several key informants said that referrals from primary to secondary care work in similar ways as referrals for other specialty medical services. Overall, most hospital providers felt the pathway into care at their hospital was functional and timely, as staff routinely make efforts to get patients in as quickly as possible. As one OB-GYN director summarized it, “there is a telephone line for them to phone us, and it all works seamlessly after that. The patient is referred to the next clinic” [Provider 43]. A few respondents felt that this referral pathway works more efficiently at some hospitals than others, and in some places it relies on established professional relationships between individual GPs and OB-GYN consultants. GPs are expected to phone, e-mail or fax an established coordinator who may be a clinician (generally a midwife) or an administrator. One OB-GYN, for example explained that at her hospital, there is “someone whose job is nine to five manning a phone line and doing the secretarial work,” which was very helpful in establishing the service in the early stages [Provider 24]. In some other hospitals, a nurse midwife holds a dedicated cell phone and hands it off to others when she is unavailable. A few providers did mention that referrals are dependent on the coordinator being available and that challenges arise when this person is on leave. One provider explained that the liaison person at her hospital “isn't great at answering the phone,” but that she returns phone calls and emails in a timely fashion [Provider 31].

3.3. Remaining challenges

3.3.1. No provision or partial provision in some hospitals

One important remaining challenge in the implementation of hospital-based abortion care is that some hospitals do not provide any TOP services or provide them only in cases of FFA but not under section 12 (early pregnancy). We interviewed two midwives from hospitals that do not provide any TOP, and two from hospitals that provide only for FFA, and learned that the primary reason why hospitals do not provide is CO among staff. One midwife from a hospital that does not provide any TOP felt that these objections came from consultant obstetricians as well as anesthesiologists and nursing staff. She was able to convince her staff to provide ancillary services such as ultrasound scans and Rho(D) immune globulin despite some initial “disquiet about this”, but no actual abortion services [Provider 47]. Another midwife in a hospital that now provides services for FFA explained that the mandate to introduce TOP was a challenge for her hospital. She negotiated at length with consultants and other staff to introduce the service, and a new consultant hire facilitated the process in the end:

“All of our doctors here were conscientious objectors, […] So the next phase was to bring in fatal fetal abnormalities. So we moved, after the fourth consultant was appointed we moved to get that service, so you bring all of the team together and see who will provide the service. […] So three of our consultants now are providing services for fatal foetal abnormalities, and really it was working out sort of pathways for them, how they would be provided, what the service would look like for women when they arrived for that service, so that's put my role in sort of, I suppose No. 1 seeing who is going to provide the service?” [Provider 51, midwife]

Another midwife explained that she has been unable to convince the consultants, and as a result her hospital is not in compliance with the national mandate and does not provide any TOP:

“The barriers that I have, I can't actively move those barriers so and I have to respect people's views. I have to, I mean, I work with these people, so I have to respect the views and how they feel. […] From a national perspective, there has been, it's been very clearly said that we are one of the outliers as such and therefore we are not conforming and […] we're not the flavor of the month, let's just say, but that really doesn't wash with the consultants. They don't care. Not that they don't care, I mean they don't care that that's a perception. That doesn't bother them, but I mean, so there have been different conversations held at a very high level but no. They haven't wavered from, that is their belief and that is what … they won't budge on it, they won't budge on it.” [provider 49, MW director]

Another common reason given for why some hospitals do not provide TOP was inadequate facilities. Some are maternity hospitals only and do not have a dedicated gynecology ward, which raises concerns about whether abortion can be provided in the same space as other obstetrical services. One midwife director in a hospital that provides only for FFA said that it seems “contradictory” and “not right to [her]” to provide abortions on a maternity ward, for example [Provider 40]. Two other midwives in a non-providing and a partially providing hospital also said that the lack of a dedicated gynecology ward had been a barrier to introducing the service in their hospitals. These midwives also cited staffing limitations as a barrier, and one of them added that people initially thought there would be no demand for abortion at her hospital.

Regardless of the reasons why abortion services were not introduced at some hospitals, respondents from non-providing hospitals felt that there is little incentive to introduce services now as the demand is being met by nearby hospitals. As one midwife explained: “I think really probably the opinion is that [nearby] hospital is providing the service so the problem has gone away” [Provider 47]. Another midwife added: “We are lucky because [there is] a hospital that's so close geographically that does provide, that is definitely a safer place for these women to go to” [Provider 40, midwifery director]. She further explained that the nearby hospital has more annual births and had more resources and staff to introduce the service. A senior obstetrician who directs several hospitals, some of which do provide TOP and some of which do not, explained that pushing back on referrals from outside hospitals had been effective in encouraging one hospital to begin providing at least for FFA:

“So, initially we did, and we kind of went along and said look at, we'll take this case but we really need you to look at taking your own cases. Our midwives are a bit unhappy taking cases from other regions, when they are only just initiating … getting going on their own for our cases and … so with a bit of good will, and a bit of cajoling, and finally … towards September/October, a final sort of well, we haven't the capacity to take your cases […] and finally the doors began to open.” [Provider 43, OB-GYN]

However, this strategy had limited success, because it worked with one but not all the hospitals under this respondents’ direction.

From the HSE's perspective, several key informants said that there are ongoing efforts to engage non-providing hospitals. One key informant explained that the north-west and south-west regions of the country are currently poorly served, but there is an ongoing effort to expand services to hospitals which are not currently providing. She added that the HSE has earmarked funding for TOP services, which some hospitals have used to purchase equipment and hire staff to provide the service, including midwives and consultants. The HSE approach to encourage non-providing hospitals to change course is to provide funding for the integration of TOP services into other women's health and ambulatory gynecology services [KI25]. This strategy introduces TOP while expanding other ambulatory gynecology services, which is beneficial across the board, as there currently are long gynecological waiting lists which this type of service would address [KI25]. At the time of data collection, the HSE had not received any such proposals (which can be relatively informal), but several respondents thought this was a promising strategy.

3.3.2. Marginalized service relying on few willing providers

Although some providers felt abortion care had become more acceptable over time and integrated into hospital services, others felt that TOP remained a marginalized service for both patients and providers. One OB-GYN explained that at her hospital, TOP patients are only seen during evening clinics. When they are admitted for medical abortion, this happens on weekends only, when the wards are less busy with other gynecological patients. The service was structured this way due to limited space.

“The clinic starts at 5:00 or 6:00pm and goes until maybe 8:00 or 9:00pm, depending on the number of patients […] So yes, for staff, it's not that great I find and then even if I come across a patient in the evening who's got, you know, some comorbidity and I need to discuss it with haematology, anaesthetics or somebody, there's no one there. And even for booking, if I need to book a surgical procedure, they're all gone, so it's just a bit limiting that way, I think. So yes, so it is supported from hospital management, but it had to fit in around things already in place. Yes. […] It's not going to involve cancelling another clinic that we run. So they kind of, they worked around it and like the clinics, space is at a premium in the hospital. It's an old hospital and there is no capacity for adding in extra clinics […] but then again, it's a bit of a stigma I think kind of coming in in the evening, that you know, you can't be, you know, it's relegated to that, not an important time of the day and people who work in these clinics, you know, their time isn't regarded as much. [Provider 31]

According to this provider, persistent abortion stigma contributed to the fact that services were structured at the margins of other clinical services. At least one other provider said the “TOP clinic” is a consultant-run clinic that happens in the evenings. Weekend admissions for medical abortion are preferred at some hospitals, and sometimes need to be scheduled around objectors and according to the availability of willing nursing and midwifery staff. Consistent with the idea that TOPs are generally scheduled strategically around other services, one anesthesiologist added that at her hospital, when surgical cases take place they need to be performed “first thing in the morning to insofar as possible limit any of the problems like […] the staffing issue with people morally objecting” [Provider 30].

Several providers – midwives, nurses and obstetricians alike – noted that the services “are hinging on a very small number of people” and are vulnerable to “collapse” in their absence [Provider 25, nurse theater manager]. In some hospitals back-up coordinators and providers are available but this is not always the case. As one midwife put it:

“The hardest thing I find is to have cover for me. Like in the past, I wanted to have a Tuesday off. I'd book a day's annual leave and I […] spoke to my colleagues, that was fine. Whereas it seems like it's only me now and it's a lot harder to get away or to finish early or to do, you know, like I'm on annual leave now […] I'll have to email, that's a job I have to do next week is email and wait and see, you know, and I have a timetable up Monday, Tuesday, Wednesday, Thursday, Friday and to see who responds, you know, because I have to arrange all that. It's not like, ‘Right, you go away on your annual leave and we'll organize it’. I have to arrange my own cover.” [Provider 13]

Many respondents cited recruiting additional willing providers as a means to solve this challenge, but this is not always feasible, despite the Health Minister's support for doing so. One nurse who is an operating room manager explained that asking potential recruits about their stance on abortion is not deemed culturally acceptable, especially on the nursing and midwifery side but sometimes even on the physician side:

“No, you can't ask that question. You can't. Yes, you can't not offer a job because somebody is a conscientious objector. It can't come into it at all. It's weird because even the consultant anesthetist who was appointed on the basis of funding for our service and he was appointed with money that was coming through for abortion services and he's actually a conscientious objector […] You can't ask. This is going back to what I was … Even on the medical side, you can't ask. Now, you would assume that somebody who's going to be providing the service on the surgical side wouldn't go for the job if they were an objector, but I can't ask a midwife or a nurse at interview if they are willing to be involved in abortion services and use that as one of my markers for whether I give them a job or not”. [Provider 25]

Other respondents did mention having successfully recruited new consultant OB-GYNs (often younger newer graduates) to provide services, especially for FFA. One OB-GYN explained that “the funding for her position came from the termination of pregnancy services, so [she] was very aware that it would be part of [her] job description”, and she was asked at her interview to confirm she would provide the service. However, such recruitment efforts have not always been successful, and an OB-GYN director who has been unable to find a willing consultant to fill a position confirmed that “we're in not in a position in Ireland to give or not give a job based on the answer to that [TOP] question” [Provider 43]. The interviewee did not explicitly explain the reason but implied that if a provider otherwise qualified for a position refuses to perform abortions it would be considered discriminatory not to offer the job only for that reason.

Peperoncino2∗

3.3.3. Lack of access to surgical abortion

Another important theme that emerged from our data was the limited access to surgical abortion. Second-trimester surgical abortion (dilation & evacuation) is not available anywhere as this is “not part of training in Ireland” [Provider 35, OB-GYN]. Even for first-trimester cases, most respondents acknowledged that hospitals that offer uterine aspiration as an initial option are rare. As one OB-GYN explained: “most people are counseled towards medical and the vast majority of surgicals are done for failed medical” [Provider 43, OB-GYN]. Surgical procedures performed in operating theatres typically involve nurses and anesthesiologists in addition to a surgical provider, and it can be challenging to find willing staff in each category. As one nurse who is a theater manager explained:

“I will generally pick a theatre based on whether I have willing participants in that room to be involved and if there aren't, then I have to find from the rest of the staff in the department who is going to be involved in the case and commonly, that will be a manager, a midwife and maybe one of the theatre scrub team because we have a lot of conscientious objectors. [Provider 25]

One OB-GYN provider who performs surgical TOP also explained that such cases only happen “once every four or six weeks”; the consultants at her hospital “purposefully try to limit them because it is a barrier trying to get the theatre staff in” [Provider 31]. One OB-GYN trainee explained that some OB-GYNs “are happy to do all forms of medical termination for all permitted reasons, but they will have a problem doing surgical termination […] because it feels more like […] a definitive act where you're ending the pregnancy rather than just giving tablets” [Provider 22].

One service user added to these data by recounting a frustrating delay in obtaining a uterine aspiration despite having experienced symptoms for an entire month after an incomplete medical abortion:

“I took the first phase of tablets on the 10^th/12^th January 2020 […] After I took the tablets, the foetus came out. But the placenta didn't […] On the Thursday, I started bleeding and getting contraction pains […] They asked me to take paracetamol. […] I asked the other consultant if I could have a D&C. But I was given 8 more Misoprostols. I took the 4 tablets on Friday and the other 4 tablets, 48 hours later […] I had a follow-up scan […] the scan revealed that I had retained some placenta. At that point I said that I wanted a D&C, but I was being offered more tablets. Then I said that I want a D&C and my consultant called me up and said that she'd do the D&C the next day. So, I finally got a D&C on 11 February.” [Service User 6]

This case suggests that at least in some instances repeat medical management remains the preferred choice even after an initial failure.

Our data show multiple barriers to routine provision of surgical abortion. Even when willing OB-GYN, anesthesia and nursing providers are available, logistical barriers remain as access to theatre is scarce in most hospitals due to long waiting lists and insufficient theatre space. As one midwifery director explained:

“Two [consultants] offer the surgery and I suppose that was a bit of a concern because, [the] gynae waiting list is so long and now I've lost a slot because of TOP or you know, we would have had problems scheduling people for miscarriages that needed ERPCs¹ and now we've given up an elective slot for a TOP. And that judgment comes in as in which is the more deserving patient? Now, people don't say that out loud, but sure that's what they mean.” [Provider 41]

An OB-GYN further explained that all obstetrical and gynecological cases share the same three theatres, which means frequent obstetrical emergencies limit access for other cases:

“[This is a very busy] maternity hospital in Ireland. […] And we've got a very high Caesarean section rate, so our theatres are chock-a-block, you know […] So you can understand the reluctance to give up any theatre space to surgical terminations, you know. We only have […] two functioning theatres and one emergency theatre.” [Provider 26, OB-GYN]

Although surgical abortion can be performed under local anesthesia in ambulatory settings, which eliminates the need for theatre and anesthesia staff, this is rarely done in Ireland according to our interviewees. One provider said that she does routinely provides manual vacuum aspiration (MVA) in an ambulatory setting, however the service was new at the time of the interview and primarily reserved for patients with retained products after a medical abortion [Provider 31]. An OB-GYN from a different hospital in another region also said MVA is routinely offered at her hospital and explained that this is because of an earlier effort to build capacity for providing gynecological procedures in an ambulatory setting. Once the abortion law came into effect integrating MVA into this space was relatively easy:

“So we had talked about bringing in MVA for miscarriage work. We hadn't got there exactly, but we had developed good pathways for women coming in for operative hysteroscopy […] So the staff in the ambulatory unit became skilled in, you know, giving support to women having operative procedures in a side room […] So I think one of our facilitating features was that all practitioners stayed as general and developed their side room skills, which meant that the nurses didn't see it as a specialized service just for one practitioner. So there was capacity in that unit every day that there were procedure lists going on. And that same unit houses the early pregnancy ultrasound service, so it was, you know, the ultrasound availability was there as well.” [Provider 36]

As mentioned above, key informants from the HSE found that developing this kind of ambulatory gynecology unit could be a solution for hospitals that are not currently providing TOP. However, descriptions of existing units of this sort were rare. Most abortions (medical or surgical) happen instead in “shared infrastructure” on gynecology or maternity wards, without any “audiovisual separation” between types of patients, which providers said can be a challenge [Provider 37, midwife]. Finally, a few providers mentioned having received training and/or MVA equipment, but this service is not routinely offered at most hospitals. The providers who received MVA training were all OB-GYN, but one midwife we interviewed said she would be “100% up for” receiving MVA training, although midwives in Ireland are not currently performing MVAs or any other kinds of abortions [Provider 37, midwife].

4. Discussion

In this analysis, we described the early phase of the implementation of hospital-based abortion services in Ireland and highlighted successes and remaining challenges. We found that individual champions are key to establishing the service, but their motivation is not always sufficient to integrate TOP into the existing clinical services. Abortion care remains peripheral and marginalized at some hospitals, in part because of limited hospital capacity, and in part because of persistent stigma around abortion care. At other hospitals, abortion services are not available at all, and the number of providing hospitals has not increased over time. Conscientious objection is a key challenge that emerged early in the implementation process and continues to limit the expansion of abortion services. Access to surgical abortion is limited, even for MVAs. Values clarification workshops emerged as helpful to clarify and sometimes expand health worker roles.

These findings echo what some previous studies of abortion implementation have found in other countries. For example, one multi-country study of abortion implementation also found that clinicians’ willingness to provide abortions is key to success, and that this willingness depends on several factors, including their knowledge of the law, their personal views about abortion, the specifics of the case such as gestational age, and on the abortion method (medical versus surgical) (Glenton et al., 2017). In Uruguay and Portugal, similar to Ireland, implementers chose to focus on medical abortion, which was logistically easier to implement and also less likely to elicit strong objections from providers (Stifani, Vilar, & Vicente, 2018; Stifani, Couto, & Lopez Gomez, 2018). A survey study of Irish OB-GYNs in training also showed that many are willing to provide medical but not surgical abortion, which they consider more “real” and “active participation” in abortion (Stifani et al., 2021). Focused clinical training initiatives may be helpful in increasing the number of providers who are comfortable and willing to provide aspiration procedures.

In addition to provider-level barriers, we found that inadequate facilities were another limitation to the expansion of surgical abortion services. Creating ambulatory gynecology units for TOP provision, in addition to other outpatient surgical procedures, may be a promising solution to this problem where achievable from a funding and logistical standpoint. Other possible solutions include expanding capacity for MVAs performed under local anesthesia without the requirement for special units or additional staff and expanding who can provide abortion related care as per WHO guidance (World Health Organization, 2016).

In Ireland as in other countries, addressing barriers to the provision of surgical abortion is key, as this is an important component of a quality abortion service. According to the WHO, quality care is care that is consistent with evidence-based professional knowledge, and is people-centered, safe, effective, timely, equitable, integrated, and efficient (World Health Organization, 2020). Providing quality, person-centered abortion care should include, wherever possible, a choice of medical versus surgical methods, as having a choice of methods is important to most women undergoing abortion, and women are more likely to find a method of abortion acceptable if they choose it themselves (Henshaw et al., 1993; Slade et al., 1998; World Health Organization, 2016).

Another important finding from this study is that although the Minister of Health was motivated to implement the service quickly, several providers perceived delays and a lack of top-down leadership from the HSE. This led to less organized beginnings for the hospital-based service and created challenges for implementers on the ground. In a study of abortion implementation in six countries, researchers found that political will is key to ensuring successful implementation (Chavkin et al., 2018). Key informants in Portugal explained that the speed at which the ministry of health had organized stakeholders following the referendum was essential in ensuring successful implementation of the new law (Stifani, Vilar, & Vicente, 2018). Thus, a lesson for other countries is that political will must be accompanied by swift action and ministerial leadership to organize guideline development and plan implementation on the ground. This preparation should begin even before the law comes into effect.

Conscientious objection was an early challenge to establishing hospital-based abortion services in Ireland and a persistent barrier to expanding them. This echoes findings from other countries, where CO limits and sometimes even eliminates access to abortion (Autorino et al., 2020). Interestingly, our respondents almost universally described values clarification workshops as helpful in defining the roles that providers can serve despite CO. Even willing providers were able to reflect on their experience and found the workshops valuable. This is a key finding because it positions such workshops as an important element of abortion policy implementation. While researchers have previously found that values clarification workshops improve knowledge, attitudes and intentions to provide abortions (Turner et al., 2018), we are not aware of other studies which have described the role of these workshops in the context of policy implementation. Another potential solution to CO is to recruit new, willing providers, clearly stating that abortion provision is expected. Data from other countries show that successful strategies exist for recruiting abortion providers (McLemore et al., 2015), and that exposure to abortion during clinical training contributes to the recruitment effort as it is a predictor of abortion provision in future practice (Steinauer et al., 2008).

While our analysis of the establishment of hospital-based services highlighted more barriers than our work on the establishment of community-based care, which was largely successful (Mishtal et al., forthcoming), most of the findings we report here agree with or complement those we reported. For example, hospital-based providers, like GPs, highlighted the importance of collaborative efforts and supportive groups of practitioners. Data from GPs also showed that where hospitals do not provide services, GPs may be hesitant to do so as well as they do not have the “backup”. Unequal distribution of services is an issue that needs to be addressed at the secondary and primary level alike. An interesting finding of this analysis is that many hospital providers and key informants alike perceive that hospitals are meeting the needs of most women who need abortions within secondary care, and that referral pathways from primary care are generally smooth and timely. However, data from GPs show that referrals to hospitals are often a challenge (Mishtal et al., forthcoming). The HSE has taken steps to improve this pathway by collecting updated contact information for each hospital's abortion service coordinator, but this will likely require ongoing efforts to ensure smooth referrals across the board.

This study has several limitations. First, service users and providers self-selected to be a part of the study and may have had particularly strong views on the topic of abortion and its implementation in Ireland, or they may have had a uniquely negative or positive experience accessing services. Although we explicitly sought to interview providers in hospitals that do not currently offer abortion services, we were not able to interview anyone who clearly identified as a conscientious objector. This may be in part because the study topic may have generated more interest among people who support increased access to abortion. Also, we had few sample users who sought care in a hospital setting, including only one user who had a hospital-based TOP under 12 weeks, limiting our ability to appreciate all factors relevant to the patient experience.

Despite its limitations, this study presents the perspectives of several types of stakeholders across Ireland on the implementation of hospital-based abortion services after legal change. As such, it adds important information to the body of knowledge of how new abortion policy is enacted on the ground. We hope that these findings will inform a review of the Irish abortion service as well as other countries that are planning to establish or expand abortion services in the future.

Ethical statement

The WHO's Ethics Review Committee approved this study (#AA6601). The University of Central Florida's Institutional Review Board (#000846) also approved this study.

Funding

The UNDP-UNFPA-UNICEF-WHO-World Bank Special Programme of Research, Development and Research Training in Human Reproduction (HRP), a co-sponsored program executed by the World Health Organization, funded this work (grant # AA6601).

Author contributions

All authors meet criteria for authorship and reviewed and approved the final version of the manuscript. Detailed author contributions as follows:

BS: methodology; analysis; investigation; writing (original draft)

JM: conceptualization; methodology; analysis; investigation; writing (review & editing); supervision; project administration.

WC: conceptualization; methodology; writing (review & editing); supervision.

KR: methodology; investigation; analysis; data curation; project administration; writing (review & editing)

LG: investigation; analysis; writing (review & editing)

DC: investigation; analysis; writing (review & editing)

DD: conceptualization; methodology; analysis; writing (review & editing)

MM: conceptualization; methodology; resources; writing (review & editing)

TH: conceptualization; methodology; resources; writing (review & editing)

MF: conceptualization; methodology; resources; writing (review & editing)

AL: conceptualization; methodology; writing (review & editing); supervision; project administration; funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.